Information on the development of involvement with alcohol between childhood and adolescence is lacking. Most longitudinal studies of alcohol use have started with adolescents and followed them into adulthood. Consequently, more is known about the natural history of alcohol involvement from adolescence on, despite recent findings showing early onset alcohol use as a risk factor for later alcohol problems. In addition, little is known about the development of risk factors for adolescent drinking because these have generally not been assessed prior to adolescence. Nor is information available regarding the influence of preadolescent risk factors on alcohol use onset and transitions into problematic alcohol use in adolescence. Such Information can have important implications for the design of alcohol prevention programs in the elementary schools. This is an application to fund the continuation into adolescence of the Teen to Teen Study, an ongoing longitudinal study of over 400 Caucasian and African-American 8- and 10-year-old children and their families, from whom seven previous waves of data will have been collected. The study is unique in its measurement of alcohol socialization influences and personality and attitudinal risk factors for later alcohol involvement. An important feature of the study is the 6-month interval between interviews, allowing us to capture rapid changes in these antecedents as well as to estimate more precisely the relation between developments in risk and drinking onset. In addition to the target children, the sample consists of their biological mothers, their biological fathers (or mother's current partner), an older same-sex sibling (if available), and one or two close friends. Retention of the sample over time has exceeded 90% through the first five waves of data collection, establishing a firm basis for the continuation of the study into adolescence when their alcohol involvement will escalate. An additional seven waves of data collection are proposed in the present application to provide nearly continuous assessment of these two cohorts of children every six months from ages 8 and 10 through ages 15 and 17, respectively.